Model comparison
GPT-4.1 vs GPT-4o
GPT-4.1 is the stronger model overall, scoring 35.9 to 28.6 on the Noometry Index.
Last verified . 47 shared benchmarks.
Summary
- They share 47 benchmarks with published results for both. GPT-4.1 scores higher in 10 categories and GPT-4o in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-4.1 leads 34.7 to 21.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 38.3% for GPT-4.1 and 6.4% for GPT-4o.
- GPT-4.1 is cheaper at $2 / $8 per million input/output tokens, against $2.50 / $10 for GPT-4o.
- GPT-4.1 accepts more context: 1.05M tokens versus 128K.
Side by side
| GPT-4.1 | GPT-4o | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 35.9 | 28.6 |
| Released | 2025-04-14 | 2024-05-13 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 128K |
| Max output | 33K | 16K |
| Input $ / M tokens | $2 | $2.50 |
| Output $ / M tokens | $8 | $10 |
| Results tracked | 52 | 72 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4.1 leads
GPT-4.1: 34.4 (#238), GPT-4o: 24.8 (#328)
| Benchmark | GPT-4.1 | GPT-4o |
|---|---|---|
| SWE-bench Verified | 48.5% | 31% |
| SWE-bench Verified (bash only) | 39.6% | 21.6% |
| Aider Polyglot | 52.4% | 45.3% |
| WeirdML | 39% | 25.1% |
| LMArena Coding | 1391 | 1297 |
| CadEval | 42% | 26% |
| GSO | — | 0% |
| BigCodeBench Instruct | — | 51.1% |
| LiveBench Coding | — | 51.4% |
| BigCodeBench Complete | — | 61.1% |
| ALE-Bench | 558.1 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |
Agentic & Tool Use GPT-4.1 leads
GPT-4.1: 34.7 (#43), GPT-4o: 21.0 (#141)
| Benchmark | GPT-4.1 | GPT-4o |
|---|---|---|
| Berkeley Function Calling Leaderboard | 54% | — |
| GDPval | — | 9.9% |
| TheAgentCompany | — | 8.6% |
| Cybench | — | 12.5% |
| BALROG | — | 32.3% |
| LMArena Search | — | 1006 |
| METR Time Horizons | — | 40.8% |
Reasoning GPT-4.1 leads
GPT-4.1: 11.7 (#339), GPT-4o: 9.4 (#343)
| Benchmark | GPT-4.1 | GPT-4o |
|---|---|---|
| ARC-AGI-2 | 0.4% | 0% |
| SimpleBench | 27% | 17.8% |
| ARC-AGI-1 | 5.5% | 4.5% |
| Chess Puzzles | 6% | 13% |
| EnigmaEval | 2.2% | 0.8% |
| LMArena Hard Prompts | 1384 | 1281 |
| DTBench | 68.3% | 64.5% |
| LMCA | 25.6% | 16.6% |
| Epoch Capabilities Index | 136.78 | 128.97 |
| ForecastBench | 61.5 | 57.7 |
| Kagi LLM Benchmark | 52.3% | — |
| CritPt | — | 0% |
| LiveBench Reasoning | — | 55.8% |
| LiveBench Data Analysis | — | 60.9% |
| LiveBench | — | 55.3% |
Math GPT-4.1 leads
GPT-4.1: 22.3 (#280), GPT-4o: 10.6 (#312)
| Benchmark | GPT-4.1 | GPT-4o |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6% | 0.4% |
| OTIS Mock AIME 2024-2025 | 38.3% | 6.4% |
| Omni-MATH | 47.1% | 29.3% |
| LMArena Math | 1370 | 1285 |
| MATH Level 5 | 83% | 53.3% |
| FrontierMath (Feb 2025 set) | 5.5% | 0.3% |
| LiveBench Math | — | 49.5% |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge GPT-4.1 leads
GPT-4.1: 37.1 (#160), GPT-4o: 28.8 (#242)
| Benchmark | GPT-4.1 | GPT-4o |
|---|---|---|
| GPQA Diamond | 66.9% | 49.2% |
| Humanity's Last Exam | 5.4% | 2.7% |
| SimpleQA Verified | 31.1% | 26% |
| MMLU-Pro | 81.1% | 71.3% |
| Vectara Hallucination Rate | 5.6% | 9.6% |
| GPQA (HELM) | 65.9% | 52% |
| LMArena Expert | 1364 | 1250 |
| Confabulations | — | 15.3% |
| MMLU | — | 88.1% |
Multimodal GPT-4.1 leads
GPT-4.1: 38.2 (#67), GPT-4o: 34.5 (#91)
| Benchmark | GPT-4.1 | GPT-4o |
|---|---|---|
| LMArena Vision | 1211 | 1137 |
| GeoBench | 72% | 71% |
| Video-MME | — | 71.9% |
| VPCT | — | 40% |
| ScienceQA | — | 88.5% |
Multilingual GPT-4.1 leads
GPT-4.1: 49.4 (#133), GPT-4o: 43.2 (#186)
| Benchmark | GPT-4.1 | GPT-4o |
|---|---|---|
| LMArena Non-English | 1370 | 1283 |
| LMArena Chinese | 1382 | 1277 |
| LMArena French | 1382 | 1304 |
| LMArena German | 1381 | 1282 |
| LMArena Japanese | 1319 | 1257 |
| LMArena Korean | 1339 | 1234 |
| LMArena Russian | 1377 | 1286 |
| LMArena Spanish | 1376 | 1292 |
Instruction Following GPT-4.1 leads
GPT-4.1: 71.3 (#153), GPT-4o: 66.6 (#207)
| Benchmark | GPT-4.1 | GPT-4o |
|---|---|---|
| IFEval | 83.8% | 81.7% |
| LMArena Instruction Following | 1367 | 1278 |
| LiveBench Instruction Following | — | 68.6% |
Long Context Too close to call
GPT-4.1: 40.0 (#163), GPT-4o: 39.4 (#179)
| Benchmark | GPT-4.1 | GPT-4o |
|---|---|---|
| Fiction.LiveBench | 63.9% | 66.7% |
| LMArena Longer Query | 1385 | 1289 |
Writing & Preference GPT-4.1 leads
GPT-4.1: 57.6 (#125), GPT-4o: 52.6 (#166)
| Benchmark | GPT-4.1 | GPT-4o |
|---|---|---|
| LMArena Text | 1383 | 1300 |
| LMArena Creative Writing | 1363 | 1292 |
| WildBench | 85.4% | 82.8% |
| LMArena Multi-Turn | 1398 | 1302 |
| Short-Story Creative Writing | — | 81.8% |
| EQ-Bench Creative Writing | 1420 | — |
| LiveBench Language | — | 47.6% |
Frequently asked questions
Is GPT-4.1 better than GPT-4o?
GPT-4.1 is the stronger model overall, scoring 35.9 to 28.6 on the Noometry Index.
Which is cheaper, GPT-4.1 or GPT-4o?
GPT-4.1 is cheaper. It lists at $2 per million input tokens and $8 per million output tokens; GPT-4o lists at $2.50 and $10.
Is GPT-4.1 or GPT-4o better for coding?
GPT-4.1 scores higher on coding benchmarks: 34.4 versus 24.8 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 128K.
How many benchmarks do GPT-4.1 and GPT-4o share?
47 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and GPT-4o has 72.